Are You Lying to Me? (Part 1)

Imagine, on a whim, you have proffered $44 billion to buy a social media platform, have a large fortune of your own at risk in the deal, and now are wondering if this was such a good idea. And you are sitting across the table from the social media executives, negotiating the deal, and trying to make sense of whether what they are telling you is true or a lie. How many bots really are on their network?  Are they not making eye contact? Are they fidgeting? Do they appear nervous? Are they stumbling over their speech a bit? Are they looking up and to the right?

Guess what? Not one of those things are relevant to whether they are lying or telling the truth. Of course they are fidgeting, stumbling, and looking away…they have a reason to be nervous, they are trying to figure out why you want to pay them 35% more than their company is worth, and they don’t want to blow it. And they are hyperaware that everything they relate to you may impact the value of their stock and thus their retirement plans.

If you negotiate deals at this scale or any scale, or if you interview people for a living, you need to be able to discern whether what they are telling you is true. The problem is, deception detection is difficult, people naturally are not very good at it, and most everything we have been taught is wrong.

Two decades of solid research has made very clear that folk methods of detecting deception are almost universally unrelated to lying. These include supposed cues such gaze aversion, personal adaptors, “nervous behaviors” like fidgeting, and micro expressions of contempt or the like. To be overly clear on this point…they tell us nothing about veracity. If you go through training on body language deception detection, you’ll be more confident going into that meeting, and more confident in the veracity assessments you make, but your accuracy will be no better…and the research actually shows it probably will be worse.

I’ve spent the past fifteen years operationalizing the work of the top researchers in the field of deception, and I am confident that what I offer can make you significantly better at detecting deception. What I’m not going to offer is an easy, quick fix. Because there isn’t one.

In this installment I want to look at theories underlying deception, and provide a glimpse into how we might detect it, and in the next installment I’ll go deeper into practical detect deception methodologies.

Professor Timothy Levine has done a fantastic job in his book Duped[1] laying out the underpinnings of lying. Most of it isn’t new but he does a great job elaborating the theory of why and how people lie. 

At the heart of Levine’s work is that most people tell the truth most of the time. When we lie, it is for a purpose. In other words, lying is strategic. If the truth won’t suit me in this context, and relative to the questions I am being asked, I strategically decide to lie. Along with this is the fact that most untruthful statements are still composed primarily of the truth. Creating lies and remembering lies takes energy and our brains are designed to conserve energy. The corresponding theory to this is that most people believe other people most of the time. Again, it takes energy to question and analyze a narrative. Taking these together, Levine provides Truth Default Theory.

Next, there are no universal cues that give away a lie. As Levine notes, “what gives away one liar is usually different from the signals that reveal the next liar. The signs also seem to be unique to the particular lie.”  I’ll carry that a step further – the context in which the lie is being told also will change the cues. Levine hints at that and we’ll discuss that further in a moment. As many researchers have pointed out, “there is no Pinocchio’s nose.” When I lie, my nose doesn’t get bigger, I don’t look one way or the other, and I don’t fidget in any particular manner and nor does anyone else.

Not to belabor the point, but I’d also invite you to think about it this way for a moment. As I ask nearly every group I work with: “If you are putting your energy toward trying to make sense of someone’s non-verbal behavior, what are you not doing?” You’re not listening. This may not matter so much if I’m sitting in the bar with my mate listening to some crazy story, but it matters incredibly in high stakes interviews and interactions such as a negotiations. We need to hear what is said, we need the cognitive space to make sense of what is being said, and we need to hear how they say it.

As a side note, since I mentioned high-stakes situations, Levine and other researchers have found no difference between high- and low-stakes situations in terms of how we lie or in terms of our behaviors when delivering the lie. Someone lying about a murder shows no more cues, and no different cues, than someone lying about having eaten the last cookie.

Still, understanding context is critical. What information is delivered and how it is delivered always changes depending on the operative context. When looking at context, there actually are two contexts we need to keep in mind. There is firstly, the context in which the events being related occurred and secondly, the goal-directed context in which the statement (lie or truth) is being delivered.

As the interviewer or listener, we need to make sense of the confluence of these contexts. Thinking back to the original example, in what context did the social media executives agree to sell the company, what information do you have about that context to provide perspective, and in what context are they now operating, i.e. what is their current motivation and is that different from the original. This will help you begin to frame their statements better and better separate truth from non-truth.

This brings us to the final two elements of Levine’s theory: correspondence and coherence.  Regarding correspondence, does the narrative provided correspond with known evidence or ground truth? Essentially he is advocating an evidence-based method of deception detection and correctly points out that if we have evidence that does not match with what the person is saying, there is a good chance that they are lying, and we can beat that 54 percent average. Fair enough.

The next aspect, coherence, is a bit trickier. Basically what Levine is saying is that we need to look at whether the narrative is internally coherent. He admits the evidence isn’t fully in support of that one, but it does relate to the overall assessment of plausibility with respect to content, context, and correspondence.

I bring Levine so centrally into this for two reasons. First, via Malcolm Gladwell’s Talking to Strangers, Levine’s Truth Default Theory has gained significant notoriety. Second, I believe Levine does a fantastic job at laying the foundation for understanding deception and understanding how we might go about detecting deception. From that perspective, his work is very much worth reading.

What I don’t think he does as well is provide interviewers with the ability to detect deception in the midst of an interview, and when you have no evidence to point you one way or another.  If you have evidence, use it and don’t be distracted by irrelevant cues or the person’s demeanor (because we usually understand demeanor incorrectly as well). Basically, in Levine’s system, and clearly this is from my perspective, there is no clear direction on how to make sense of whether it is more probable the person is lying or telling the truth while the conversation is occurring, which is precisely when we do need to be able to make sense of it.

While trying to make sense of plausibility and context may in the big scheme be a reasonable approach, I feel this still imposes too much cognitive load on the interviewer and detracts from our ability to really focus on content.

Next installment we’ll begin to look at another way forward, that allows us to focus on content while providing some specific verbal credibility indicators to look for within that content. I hope you find this helpful and please reach out if questions relative to what I can do to help your team excel.

[1] Not to be confused with Prof. Saul Kassin’s contemporaneous book of the same name and on a similar topic- seriously guys?